Normal Distribution Based Similarity Profiled Temporal Association Pattern Mining (N-SPAMINE)
Vangipuram Radhakrishna, Puligadda Veereswara Kumar, V. Janaki · RePEc: Research Papers in Economics · 2017
Temporal patterns in time stamped temporal databases are sequences of support values and hence, they are represented as vectors. This makes it challenging to obtain similar association patterns in context of time stamped temporal databases whose support trends change similar to a reference support sequence trend. The main idea of this work is to study the possibility of applying normal distribution concept to find similarly varying temporal patterns. This paper introduces a new approach, called N-SPAMINE for mining similarity profiled temporal association patterns by applying normal distribution which is first of its kind of approach for finding similar association patterns and uses the novel dissimilarity measure for obtaining dissimilarity between chosen temporal pattern and the reference. The results show that the proposed approach is correct and complete.